EMPIRICAL EVIDENCE FOR THE

QUANTITY THEORY OF MONEY:

ROMANIA – A CASE STUDY
PhD Candidate Alexandru PĂTUŢI
PhD Candidate Alina TĂTULESCU
Academy of Economic Studies

Abstract

The quantity theory of money is one of the most important theoretical

propositions in monetary economic theory. However, in recent days more and
more economists contest its validity. In this article we will attempt to offer
additional empirical support for the quantity theory of money. We argue that
the variation in the consumer price index (CPI) in Romania in the last six
years can be fully explained by the variation in the supply of money.
Key words: the quantity theory of money; money supply; regression
analysis; consumer price index (CPI); purchasing power.

1. Introduction

The quantity theory of money is probably one of the oldest theoretical

propositions in modern economic thought. Some of its first formulations, at
the hands of the English philosopher John Locke [1], date back to the 17th
century. At the core of the quantitative theory of money lies the proposition
that the price of money (i.e. the purchasing power of the monetary unit) is
determined by the interplay of supply and demand for money. As L. von
Mises [3] correctly points out: “This theory is essentially an application of the
general theory of supply and demand to the special instance of money”.
However, historically the quantity theory of money had one major
drawback, at least with regard to one of its more “primitive” formulations.
Numerous economists interpreted that a change in the supply/demand for
money would cause a proportional change in the level of prices1. Probably,

1. It is true that there are examples of illustrious economists who rejected the alleged
proportionality between movements in the monetary equilibrium on the one hand and prices on
the other. Richard Cantillon’s “Essai sur la Nature du Commerce en Général” [2], published
in 1755 clearly argued that an increase in the stock of money does not cause a proportional
increase in all the prices of good and services, but that the structure of prices will be modified
altogether. In his honor, the redistribution effects associated to an increase in the money supply
are known today as “the Cantillon effects”.

12 Romanian Statistical Review nr. 11 / 2013

the most famous example in this sense is Irvin Fisher’s renowned equation of
exchange M V = P T 1. This type of reasoning, which we consider erroneous,
is probably the consequence of an endeavor to study phenomena only at an
aggregate level and ignore the concrete actions of individual agents.
It was only until the first half of the 20th century when modern economic
monetary theory correctly reincorporated the core of the quantity theory of
money, i.e. the existence of a causal connection between the interplay of demand
and supply of money and the purchasing power of the monetary unit2. There was
only one essential difference. Unlike the previous versions, the focus was moved
from an alleged proportional change in the “level of prices” to a disproportional
change in relative prices. In this sense, a change in the relation between the
supply and demand for money would cause a modification in the whole structure
of prices. For example, it is true that an increase in the stock of money causes an
increase in all prices, but not all the prices rise to the same extent3.
These economic theories are, of course, a priori propositions which
do not require any additional validation. However, we consider that there was
a lack of attempts to illustrate the theory based on empirical data collected
from Romania. Thus, in this article we will use statistical data to see whether
additional support can be given to the quantity theory of money. Our main
goal will be to show that the variation in the CPI, which is used as a yardstick
to gauge inflation, can be explained by the variation in the stock of money.
In order to achieve our above stated goal, we will employ regression
analysis. The data which we will use was collected from official sources
like the Romanian National Institute of Statistics and the National Bank of
Romania.

2. Inflation and the quantity theory of money

The quantity theory of money claims that there is a causal relation

between a ceteris paribus increase/decrease in the supply of money and a
change in the purchasing power of the monetary unit. Thus, the normal
1. Where M represents the total stock of money, V represents the so called “velocity of money”,
P stands for the general level of prices and T for the number of transactions.
2. See for example the works of Mises [4] and Hayek [5]. It is interesting that numerous
economists today reject the idea that a causal connection between the purchasing power and the
supply and demand for money exists.
3. It was Mises [4] and Hayek [5] who argued that an increase in the supply of money, via
productive credit, has a disproportional effect on prices. In their view, an artificial increase
in productive credit causes a higher increase in the prices of producer goods relative to the
prices of consumer goods. This insight was used by the above mentioned authors to shape their
renowned business fluctuations theory.

Revista Română de Statistică nr. 11 / 2013 13

conclusion is that an increase in the stock of money will determine a decrease
in the purchasing power of the monetary unit and a corresponding1 increase in
prices of goods and services. It is clear that according to the quantity theory of
money, inflation is strictly a monetary phenomenon.
In order to illustrate this conclusion we have chosen to apply regression
analysis to see to what extent the variation in the CPI can be explained by the
variation in the intermediary money mass (M2).
In Romania, according to the National Institute of Statistics, inflation
is calculated based on the Consumer Price Index [6]. Thus, the rate of inflation
is computed according to the formula CPI – 100.

2.1 Methodological note

The data set is composed of monthly values in Romania between
January 2008 and September 2013. In order to describe the evolution of the
money stock we have employed data made available by the National Bank
of Romania. The chosen indicator was M2 (the intermediary money stock)
which comprises the monetary base (M1) plus demand deposits2. Because the
data was given in absolute terms (millions of lei), we chose to express it as
indexes having as reference point the prices in January 2008, according to the
formula M2 in the current month / M2 in January 2008 × 100. Table 1 shows
the transformation from absolute values to indexes.
The second set of data is composed of CPI’s collected from the National
Institute of Statistics. The indexes are calculated in the same fashion, having as
reference the month January 2008. Table 1 illustrates all the available data.

M2 and CPI
Table 1
Intermediary money stock (M2)
Date M2 Indices CPI
(thousand lei)
Sep. 2013 231,258,651.6 156.83 130.22
Aug. 2013 229,631,996.2 155.73 130.96
Jul. 2013 225,700,118.5 153.06 131.23
Jun. 2013 227,563,263.3 154.32 131.67
May. 2013 225,821,616.5 153.14 131.66
Apr. 2013 225,547,340.1 152.96 131.36
Mar. 2013 225,111,160.4 152.66 131.23
Feb. 2013 219,301,444.9 148.72 131.18
Jan. 2013 219,147,477.5 148.62 130.74
Dec. 2012 221,829,585.8 150.44 129.01
Nov. 2012 220,506,477.0 149.54 128.23
Oct. 2012 220,230,597.1 149.35 128.18
Sep. 2012 220,774,195.5 149.72 127.81

1. But not necessarily proportional.

2. NBR Glossary [7].

14 Romanian Statistical Review nr. 11 / 2013

 Aug. 2012 220,022,033.7 149.21 126.33
Jul. 2012 221,067,093.9 149.92 125.69
Jun. 2012 216,449,666.0 146.79 124.96
May. 2012 218,572,694.0 148.23 125.01
Apr. 2012 216,330,914.2 146.71 124.76
Mar. 2012 214,288,680.5 145.32 124.68
Feb. 2012 213,529,316.6 144.81 124.16
Jan. 2012 212,438,910.7 144.07 123.37
Dec. 2011 212,058,932.7 143.81 122.93
Nov. 2011 205,061,031.0 139.06 122.64
Oct. 2011 203,293,125.2 137.87 122.13
Sep. 2011 204,772,154.4 138.87 121.35
Aug. 2011 200,475,279.5 135.95 121.60
Jul. 2011 199,479,568.3 135.28 122.03
Jun. 2011 196,089,551.3 132.98 122.46
May. 2011 194,621,235.8 131.98 122.81
Apr. 2011 192,978,965.3 130.87 122.55
Mar. 2011 192,901,079.1 130.82 121.75
Feb. 2011 194,801,032.2 132.11 121.03
Jan. 2011 196,007,997.5 132.92 120.10
Dec. 2010 199,572,050.7 135.34 119.18
Nov. 2010 194,198,200.9 131.70 118.56
Oct. 2010 191,704,036.6 130.01 117.94
Sep. 2010 192,590,356.1 130.61 117.30
Aug. 2010 192,677,069.6 130.67 116.65
Jul. 2010 190,772,964.2 129.37 116.38
Jun. 2010 192,278,793.7 130.40 113.46
May. 2010 190,109,290.1 128.92 113.28
Apr. 2010 188,254,277.7 127.67 113.12
Mar. 2010 187,820,575.7 127.37 112.72
Feb. 2010 185,677,151.7 125.92 112.48
Jan. 2010 184,278,386.8 124.97 112.25
Dec. 2009 188,013,003.5 127.50 110.40
Nov. 2009 184,057,637.9 124.82 110.05
Oct. 2009 182,564,198.4 123.81 109.32
Sep. 2009 182,531,764.7 123.79 108.85
Aug. 2009 182,785,263.3 123.96 108.43
Jul. 2009 180,372,955.9 122.32 108.63
Jun. 2009 179,481,958.3 121.72 108.70
May. 2009 176,620,828.5 119.78 108.49
Apr. 2009 175,808,287.1 119.23 108.48
Mar. 2009 174,881,688.3 118.60 108.18
Feb. 2009 175,838,113.1 119.25 107.64
Jan. 2009 175,769,982.6 119.20 106.71
Dec. 2008 173,628,814.6 117.75 105.40
Nov. 2008 164,370,148.3 111.47 105.16
Oct. 2008 162,147,537.0 109.96 104.82
Sep. 2008 166,012,915.1 112.58 103.72
Aug. 2008 162,279,874.3 110.05 103.31
Jul. 2008 161,220,678.7 109.33 103.40
Jun. 2008 161,462,978.7 109.50 102.69
May. 2008 157,568,259.2 106.86 102.40
Apr. 2008 157,044,736.4 106.50 101.91
Mar. 2008 151,794,126.0 102.94 101.38
Feb. 2008 149,685,164.6 101.51 100.70
Jan. 2008 147,457,999.5 100.00 100.00
Source: [6], [8].

Revista Română de Statistică nr. 11 / 2013 15

 2.2 Trend Comparison
Before proceeding to the actual regression analysis, it is useful to
examine the evolutions using a standard graphical representation. Figure 1
illustrates the trends corresponding to the two data sets collected between
January 2008 and September 2013. There is no further need to point out that
the two variables have had remarkably similar evolutions. As it can easily be
observed, both variables had ascending trends, with M2 registering a higher
increase in the analyzed period.
At the end of the analyzed period, in September 2013, the intermediary
money stock was more than one and half times the size it represented in
January 20081. This translates, in absolute values, in an increase of 83.800.652
thousand lei over a period of approximately 6 years.
As it was mentioned above, the Consumer Price Indices registered a
similar ascending trend, increasing with approximately 30.2% over the same
period.
Summing up, in the last month of the analyzed period (September
2013), the two variables, M2 and CPI, registered values that were 56.8%,
respectively 30.2% higher than in January 2008. Thus, it can easily be noticed
that the CPI increased throughout this period at a slower rate than M2.

M2 and CPI Trends Compared

Figure 1

Source: [6], [8]

1. Exactly 156.8%.

16 Romanian Statistical Review nr. 11 / 2013

 2.3 Regression analysis
Given the fact that we’ve described the way in which the data was
collected and belabored on, we can now proceed to apply linear regression
analysis. The X variable in our case will be the intermediary money stock (M2),
while the Y variable is represented by the CPI. The reason for our endeavor
is to see whether the variation in CPI can be explained by the variation in the
money stock. The Microsoft Excel summary output, calculated based on a 95
percent confidence interval, is presented in Figure 2 below.

Regression Summary Output

Figure 2

SUMMARY OUTPUT

Regression Statistics

Multiple R 0,972519

R Square 0,945793
Adjusted R
Square 0,944971

Standard Error 2,277128

Observations 68

ANOVA

df SS MS F Significance F

Regression 1 5971,114 5971,114 1151,544 1,6835E-43

Residual 66 342,2305 5,18531

Total 67 6313,344

Standard Upper Lower Upper

Coefficients Error t Stat P-value Lower 95% 95% 95,0% 95,0%

Intercept 34,65093 2,438796 14,20821 9,4E-22 29,78171581 39,52014 29,78172 39,52014

M2 0,626057 0,018449 33,93441 1,68E-43 0,589222348 0,662892 0,589222 0,662892

Source: own calculations.

As Figure 2 informs us, there is a very strong relationship between the

independent and the dependent value. The coefficient of determination (R2)
has an approximate value of 0.95, which means that 95 percent of the variation
in the CPI is explained by the variation in the intermediary money stock.
Adjusted R Square also has an interestingly high value, which reinforces the
conclusion that the connection between the two variables cannot be attributed
to chance.
Going further, ANOVA is not extremely important for interpreting a
simple linear regression, so we will only look at Significance F to see whether

Revista Română de Statistică nr. 11 / 2013 17

our results are statistically significant or not. Usually Significance F should be
below 0.05 for the result to be significant, but that does not prove a problem in
our case because Significance F is practically 01.
The regression equation can be written as yi = 34.65093+0.626057xi.
As we can see, the P-values are practically zero, which means that there is a
zero probability that the coeficient was obtained by chance. The residual plot,
presented in Figure no. 3 below, reinforces this conclusion. We can observe
that there is no apparent pattern in the residual values, which appear to be
somewhat normally distributed and concentrated around zero.

Residuals Plot
Figure 3

Source: own calculation

2.4 Qualitative interpretation

One can easily observe that it is not hard to find empirical support for
the quantity theory of money. Economic theory clearly states that inflation is
a monetary phenomenon and that any ceteris paribus increase in the supply of
money and credit necessarily leads to a decrease in the purchasing power of the
monetary unit. The present article is only meant to illustrate the conclusion of
the quantity theory of money in a particular situation, i.e. the case of Romania
between January 2008 and September 2013.
It is clear that the relatively low inflation rates that Romania has
experienced in the past years are a consequence of the decision taken by the
NBR to keep the money supply relatively constant2. According to the quantity
theory, the only way to fight inflation is by having a fix monetary supply. It is

1. If you turn the exponential expression 1,6835E-43 into a number Significance F equals
approximately 17×10-44.
2. Although it is somewhat misleading to call an approximately 40% increase in the
intermediary money stock over a period of 5 years and 9 months “constant”.

18 Romanian Statistical Review nr. 11 / 2013

our belief that the National Bank of Romania should continue and intensify
its efforts in keeping a constant supply of money in order to prevent further
inflation.

3. Conclusions
The quantity theory of money, as we have mentioned before, is an a
priori proposition which does not require any additional validation. However,
it is interesting to point out that the theory can easily be illustrated using
statistical analysis. In Romania, in the time period between roughly 2008 and
2013, the variation in the CPI can be statistically explained by the variation
in the supply of money. Thus, the best way to fight the monetary phenomenon
known as inflation is to keep a constant stock of money.

Selective Bibliography

[1] ***Encyclopedia Britannica, 2013, available at: http://www.britannica.com/

EBchecked/topic/486147/quantity-theory-of-money. [Accessed on 10.12.2013].
[2] Cantillon, R., An Essay on the Nature of Commerce. London: Frank Cass and Co.,
1959.
[3] Mises, L. v., Human Action: A Treatise on Economics. The Scholar’s Edition ed.
Auburn: Alabama: The Ludwig von Mises Institute, 1998, p. 405
[4] Mises, L. v., The Theory of Money and Credit. New Haven: Yale University Press,
1953.
[5] Hayek, F. A. v., Prices and Production and Other Works. Auburn:Alabama: The
Ludwig von Mises Institute, 2008.
[6] ***National Instiute Of Statistics, 2013, available at: https://statistici.insse.ro/
shop/?page=ipc1&lang=ro [accessed 11.12.2013].
[7] ***The National Bank of Romania, 2013, available at: http://www.bnro.ro/
Glosar-2444.aspx, [accesed 20.12.2013].
[8] ***The National Bank of Romania, 2013, available at: http://www.bnr.ro/Masa-
monetara-M3-si-contrapartida-acesteia-5171.aspx, [accessed 20.12.2013].

Revista Română de Statistică nr. 11 / 2013 19